Search

Aim: To undertake a cost‐effectiveness analysis of dapagliflozin in treating high‐risk patients with type 2 diabetes mellitus (T2DM), using both directly observed events in the DECLARE‐TIMI 58 trial and surrogate risk factors to predict endpoints not captured within the trial. Methods: An established T2DM model was adapted to integrate survival curves derived from the DECLARE‐TIMI 58 trial, and extrapolated over a lifetime for all‐cause mortality, hospitalization for heart failure, stroke, myocardial infarction, hospitalization for unstable angina, and end‐stage kidney disease. The economic analysis considered the overall DECLARE trial population, as well as reported patient subgroups. Total and incremental costs, life‐years and quality‐adjusted life‐years associated with dapagliflozin versus placebo were estimated from the perspective of the UK healthcare payer. Results: In the UK setting, treatment with dapagliflozin compared to placebo was estimated to be dominant, with an expected increase in quality‐adjusted life‐years from 10.43 to 10.48 (+0.06) and a reduction in lifetime total costs from £39 451 to £36 899 (−£2552). Across all patient subgroups, dapagliflozin was estimated to be dominant, with the greatest absolute benefit in the prior heart failure subgroup (incremental lifetime costs −£4150 and quality‐adjusted life‐years +0.11). Conclusions: The results of this study demonstrate that dapagliflozin compared to placebo appears to be cost‐effective, when considering evidence reported from the DECLARE‐TIMI 58 trial, at established UK willingness‐to‐pay thresholds. The findings highlight the potential of dapagliflozin to have a meaningful impact in reducing the economic burden of T2DM and its associated complications across a broad T2DM population. [ABSTRACT FROM AUTHOR]

Aims: Identification, a priori, of those at high risk of progression from pre-diabetes to diabetes may enable targeted delivery of interventional programmes while avoiding the burden of prevention and treatment in those at low risk. We studied whether the use of a machine-learning model can improve the prediction of incident diabetes utilizing patient data from electronic medical records., Methods: A machine-learning model predicting the progression from pre-diabetes to diabetes was developed using a gradient boosted trees model. The model was trained on data from The Health Improvement Network (THIN) database cohort, internally validated on THIN data not used for training, and externally validated on the Canadian AppleTree and the Israeli Maccabi Health Services (MHS) data sets. The model's predictive ability was compared with that of a logistic-regression model within each data set., Results: A cohort of 852 454 individuals with pre-diabetes (glucose ≥ 100 mg/dL and/or HbA1c ≥ 5.7) was used for model training including 4.9 million time points using 900 features. The full model was eventually implemented using 69 variables, generated from 11 basic signals. The machine-learning model demonstrated superiority over the logistic-regression model, which was maintained at all sensitivity levels - comparing AUC [95% CI] between the models; in the THIN data set (0.865 [0.860,0.869] vs 0.778 [0.773,0.784] P < .05), the AppleTree data set (0.907 [0.896, 0.919] vs 0.880 [0.867, 0.894] P < .05) and the MHS data set (0.925 [0.923, 0.927] vs 0.876 [0.872, 0.879] P < .05)., Conclusions: Machine-learning models preserve their performance across populations in diabetes prediction, and can be integrated into large clinical systems, leading to judicious selection of persons for interventional programmes., (© 2020 John Wiley & Sons, Ltd.)